Images are AI-generated and illustrative.
Magnesium bisglycinate, L-threonate and taurate are three different magnesium salts. By weight they hold about 14%, 8% and 9% magnesium, so equal milligrams of compound are not equal milligrams of magnesium. Published human research also differs by form: small or short trials exist for bisglycinate and L-threonate, and we found one uncontrolled report for taurate. We found no human study that compares the three.
We sell one three-form capsule, Satiovex Tri-Mag Night: magnesium bisglycinate, magnesium L-threonate and magnesium taurate, 200 mg of elemental magnesium per serving of three capsules, 90 capsules, 30 servings. The product page lists no magnesium oxide and no magnesium citrate. Read this as a seller's account of what sources found. It makes no claim about what any form, or any capsule we sell, does.
What are the three forms made of?
Each is magnesium bound to a partner molecule: glycine (bisglycinate, also sold as glycinate; see magnesium glycinate vs bisglycinate), taurine (taurate) or L-threonic acid (L-threonate), which EFSA says arises when vitamin C breaks down. Only part of each compound's weight is magnesium.
| Form | Formula | Molar mass (g/mol) | Magnesium by weight | Compound for 200 mg magnesium |
|---|---|---|---|---|
| Magnesium bisglycinate | C4H8MgN2O4 | 172.42 | 14.1% | 1,419 mg |
| Magnesium taurate | C4H12MgN2O6S2 | 272.57 | 8.9% | 2,243 mg |
| Magnesium L-threonate | C8H14MgO10 | 294.50 | 8.3% | 2,423 mg |
| Magnesium L-threonate monohydrate | C8H16MgO11 | 312.51 | 7.8% | 2,572 mg |
The working: magnesium share = 24.305 ÷ molar mass, with IUPAC atomic weights (H 1.008, C 12.011, N 14.007, O 15.999, S 32.06, Mg 24.305). Bisglycinate: 48.044 + 8.064 + 24.305 + 28.014 + 63.996 = 172.423; 24.305 ÷ 172.423 = 14.1%. Taurate: 48.044 + 12.096 + 24.305 + 28.014 + 95.994 + 64.12 = 272.573, so 8.9%. L-threonate: 96.088 + 14.112 + 24.305 + 159.99 = 294.495, so 8.3%; one added water (18.015) gives 312.51 and 7.8%. Compound for 200 mg of magnesium = 200 ÷ share, so 200 ÷ 0.14096 = 1,419 mg of bisglycinate. PubChem lists the same masses to rounding, and EFSA's 2024 opinion gives the L-threonate ingredient as the monohydrate, specified at 7.2% to 8.3% magnesium. Magnesium acetyl taurate, seen in some animal studies, is a different compound (about 6.8% magnesium). The full table of ten forms uses the same method.
What has been published on each form?
PubMed, searched on 10 October 2026 by compound name, returned 20 records for glycinate or bisglycinate, 21 for taurate and 52 for L-threonate. The NIH Office of Dietary Supplements fact sheet (updated January 2026) names oxide, citrate, chloride, aspartate, lactate and sulfate when it compares absorption, and none of these three forms. The table lists the human studies that say most, with their limits.
| Form | Study | People, design and measure | Funding and limits |
|---|---|---|---|
| Bisglycinate | Schuette et al., JPEN, 1994 | 12 people after ileal resection; double-blind crossover; 100 mg labelled magnesium as diglycinate or oxide; absorption | US government grants. One dose, one patient group; whole-group absorption was similar (23.5% and 22.8%) |
| Bisglycinate | Schuster et al., Nat Sci Sleep, 2025 (PMID 40918053) | 155 adults with poor sleep; placebo-controlled; 250 mg magnesium a day for four weeks; Insomnia Severity Index | University institute funding; one author's firm is paid by supplement companies. Self-report only; the authors call the effect small (d = 0.2, p = 0.049) |
| L-threonate | Liu et al., J Alzheimers Dis, 2016 (PMID 26519439) | 51 adults aged 50 to 70 with cognitive complaints (44 analysed); placebo-controlled; 12 weeks; cognitive test composite | An ingredient company funded it and wrote the paper. One site; the authors say sleep and anxiety effects could not be determined |
| L-threonate | Hausenblas et al., Sleep Med X, 2024 (PMID 39252819); Lopresti and Smith, Front Nutr, 2025 (PMID 41601871) | 80 and 100 adults with sleep complaints; placebo-controlled; three and six weeks; questionnaires, cognitive tests, wearable ring | An ingredient company funded each. 2024: main questionnaire showed no group difference; registered after the study ended. 2025: both groups improved; several outcomes showed no group difference |
| Taurate | Dereli, Magnes Res, 2026 | 121 adults with elevated blood pressure; one group, no placebo; taurate plus potassium citrate with lifestyle advice for three months; blood pressure and arterial stiffness | Funding not stated in the abstract. No control group; the author says causal inference was limited |
EFSA reviewed L-threonate for the EU novel food decision (2024). It assessed safety and bioavailability, not efficacy, and noted the low number of subjects in the 2016 trial. For absorption it relied on a dissociation study, two rat studies and one human trial, in which, as EFSA summarises it, urinary magnesium was higher than placebo at week 12, plasma magnesium at week 6 but not week 12, and red-cell magnesium at neither.
All four L-threonate trials aimed at cognition or sleep were funded by, or used product from, an ingredient company, and the 2022 one (Zhang et al., PMID 36558392; 109 adults, one month, memory tests) tested a multi-ingredient capsule. We found no systematic review of them, and none of the human trials we read measured brain magnesium; brain-delivery data are from rodents. A publicly funded trial in 109 people after breast cancer surgery reported no significant difference on its pain, anxiety, depression, sleep or cognition scales (Ni et al., 2023, PMID 37520407).
For bisglycinate, absorption results in healthy people are mixed: a 2024 company-funded single-dose crossover in 40 volunteers (PMID 39770988) reported no significant rise in plasma magnesium after bisglycinate over six hours, while oxide and citrate rose at some points. For taurate, PubMed shows one study in people (above), plus rat and cell studies, reviews and hypothesis papers. We found no randomised or controlled trial of magnesium taurate, and ClinicalTrials.gov listed no study for it on 10 October 2026.
Has anyone compared the three head to head?
Not that we found. PubMed searches combining the names return no human study of any two of them, and two trial teams above say comparisons between forms are still needed. The nearest are animal studies that include magnesium acetyl taurate: a 2019 mouse study of four forms and a 2026 rat study against L-threonate. Any claim that one form is better therefore rests on different studies, with different people, doses and measures. We also found no study of any three-form capsule, ours included.
What does "taken together" mean on a label?
A label gives one figure for magnesium, and it is the element: NIH says the Supplement Facts panel declares elemental magnesium, not the weight of the whole compound. In the EU, Annex XIII of Regulation (EU) No 1169/2011 sets the reference value at 375 mg, and the label percentage is calculated against it. When one product holds two or three forms, each adds its own share and the label's figure is the total.
An example, not our label: 1,000 mg bisglycinate × 14.1% = 141 mg, 500 mg L-threonate × 8.3% = 41.5 mg and 200 mg taurate × 8.9% = 17.8 mg, about 200 mg of magnesium from 1,700 mg of compounds. A claim of 300 mg from those weights would not add up: another source, such as oxide (60.3% magnesium), would be in the blend.
If you use more than one product, add their magnesium figures. Upper reference points for supplemental magnesium differ by body: 250 mg a day from the EU Scientific Committee on Food (2001, restated by EFSA in 2024), and 350 mg a day from the US Food and Nutrition Board, which NIH says counts supplements and medicines, not food. We give no advice on combining products. NIH says impaired kidney function raises the risk of magnesium toxicity and lists interacting medicines, including oral bisphosphonates and tetracycline and quinolone antibiotics. Ask a pharmacist or doctor.
How do you read a three-form label?
- Magnesium per serving, and how many capsules make a serving.
- The forms, named in full: the share of magnesium differs by form and by hydrate.
- A weight for each form, or one "blend" figure, which blocks the ceiling test below.
- The percentage of the reference value, which refers to magnesium, not compound weight.
- Other ingredients, and any other magnesium source such as oxide.
The ceiling test: multiply each listed form weight by its share from the table. If the label's magnesium figure is higher than the sum, something else supplies magnesium; a slightly lower figure can come from hydrates. For how blends are labelled, see triple magnesium complex vs magnesium glycinate.
For magnesium L-threonate, EU Implementing Regulation 2024/2694 requires supplements to say they are for adults only, excluding pregnant and lactating women, and the Great Britain register entry NOVEL-166 requires a statement that they should not be consumed by under-18s, pregnant women or lactating women.
Our label as the example
On the label facts for Satiovex Tri-Mag Night, the magnesium line is 200 mg per serving of three capsules, 53% of the 375 mg EU reference value (200 ÷ 375 = 53.3%). The three forms are named, and the product page lists no magnesium oxide and no magnesium citrate. By the table, 200 mg of magnesium from bisglycinate alone needs 1,419 mg of compound, and from anhydrous L-threonate alone 2,423 mg, so three capsules hold at least about 1.4 g of compounds whatever the split.
The 90-capsule bottle with 30 servings has its own page. Delivery and returns are covered by the shipping policy and the refund policy, and orders carry a 30-day money-back guarantee.
Questions about these three forms
Is magnesium bisglycinate better than L-threonate or taurate?
No source we opened ranks them, and we found no human study that compares them. They differ in chemistry (14.1%, 8.3% and 8.9% magnesium by weight), in how much has been studied and, for L-threonate, in label rules. Which suits a person is a question for a pharmacist or doctor.
Can you take magnesium L-threonate and glycinate together?
We give no dosing or timing advice. Each label declares its own magnesium figure, so add them and compare the total with the reference points above. Ask a pharmacist or doctor first if you take medicines or have kidney problems.
How much magnesium is in 1,000 mg of each form?
About 141 mg from bisglycinate, 89 mg from taurate, 83 mg from anhydrous L-threonate and 78 mg from the monohydrate: 1,000 mg times the share in the table.
Does magnesium taurate have human trials?
One: a 2026 single-arm study of a taurate and potassium citrate product, with no control group. We found no randomised trial of magnesium taurate and nothing listed on ClinicalTrials.gov.
How we checked this page
- Sources opened. On 10 October 2026 we read the sources linked below, the PubMed records or full texts of the other studies named, PubChem records and the IUPAC atomic weight table. The NIH page blocks automated access, so we read an Internet Archive copy of 7 October 2026.
- What we could not find. A human study comparing the three forms, a controlled human trial of taurate, brain magnesium measured in the L-threonate trials, and a systematic review of those trials.
- Review. No doctor, pharmacist or dietitian has reviewed this page.
- Who we are. Satiovex products are sold by Xinop LLC, trading as Satiovex. We run this site and have a financial interest in the products.
- Corrections. If something here is wrong, contact us and we will check it.
Last updated 10 October 2026 by the Satiovex editorial team.
Sources
- EFSA NDA Panel, safety of magnesium L-threonate as a novel food, EFSA Journal 2024;22(3):e8656.
- NIH Office of Dietary Supplements, Magnesium: Health Professional Fact Sheet (updated 6 January 2026).
- Commission Implementing Regulation (EU) 2024/2694.
- UK Food Standards Agency, Great Britain register, entry NOVEL-166 (status date 5 March 2026).
- Schuette SA et al., JPEN 1994;18(5):430-435.
- Dereli S, Magnesium Research 2026;38(4).
- Also read, by PubMed identifier: 40918053, 26519439, 36558392, 39252819 (with erratum 40567408), 41601871, 37520407, 39770988, 30761462 and 42084749. Also read: PubChem compounds 84645, 13343447, 45489777, 76963881 and 11703161; IUPAC CIAAW atomic weights 2024; Regulation (EU) No 1169/2011, Annex XIII.